80 results about "Surface geometry" patented technology

The invention discloses a cross-source data three-dimensional reconstruction method and system based on improved Gaussian sputtering, and the method comprises the steps: collecting an unmanned plane inclined image and a ground panoramic image of a target region, and constructing a time-space correlation data set; based on multi-view geometric constraints, space-time coding matching point pairs are established through an adaptive feature pyramid, intelligent incremental cross-source data sparse reconstruction is carried out, and point cloud and camera parameters are output; adopting improved Gaussian sputtering, compressing a three-dimensional Gaussian kernel into a two-dimensional Gaussian primitive through double tangent vector constraint, and fitting surface geometry to realize multi-scale reconstruction; and optimizing primitive parameters by using a differentiatable renderer, completing multi-scale fine reconstruction through gradient back propagation, and generating a high-precision three-dimensional model. According to the method, multi-scale accurate geometric prior input and accurate camera poses are provided for three-dimensional reconstruction, the dependence on professional manual operation in a traditional three-dimensional reconstruction method is greatly reduced, and meanwhile, the geometric accuracy and visual fidelity of a reconstruction result are remarkably improved.

The invention provides a cylinder processing cutter path adaptive optimization generation method, which belongs to the technical field of cylinder processing, and adopts fractal dimension analysis to identify surface geometric complexity and determine a minimum repetitive unit, and applies a four-color theorem to carry out non-intersection area division on the surface, so as to improve the processing precision of the cylinder processing cutter path. Multi-axis linkage machining parameters are initialized, global path planning and cutter axis vector optimization and collision detection are performed based on an improved Di jkstra algorithm, path adaptive adjustment is performed according to local curvature change of a minimum fractal matrix, a game model is established, and coevolution of a double-layer game model is realized by adopting a cross optimization rule. And finally, an optimized tool path NC code containing the position coordinates, the attitude angle and the feeding speed is generated, and the technical problem that the machining efficiency and the surface quality are difficult to consider in tool path planning under the complex geometrical characteristics of the cylinder surface is solved.

The invention provides an injection molding part surface flaw detection method and device based on double-model collaboration, and the method comprises the steps: carrying out the local abnormal reflection feature analysis of a multi-view optical image collected on the surface of an injection molding part based on a first detection model, and obtaining a candidate flaw region; performing spatial positioning in the multi-view optical image based on the candidate flaw area to obtain a multi-view candidate image segment; performing surface geometric continuity analysis on the multi-view candidate image segments based on a second detection model to obtain a three-dimensional surface geometric consistency result; carrying out authenticity discrimination on the candidate flaw area based on a three-dimensional surface geometric consistency result to obtain a real flaw area; wherein the first detection model is used for capturing an optical response model of local abnormal reflection characteristics; the second detection model is a stereoscopic vision model for analyzing the geometric continuity of the multi-view lower surface. According to the invention, the false alarm rate of surface defect detection of the injection molded part under a complex surface condition is reduced.

The invention belongs to the technical field of vision measurement, and discloses a high-reflection curved surface reconstruction method and system based on multi-view point cloud fusion and optical optimization. The method comprises the following steps: acquiring multi-source heterogeneous data such as a stripe image, a polarization image and anchor point depth, and mapping the multi-source heterogeneous data to a unified workpiece coordinate system to form a standardized data set; calculating the normal direction of a sampling point based on data, establishing a slope field, and recovering the height through integration to obtain a geometric initial description of a curved surface; the method comprises the following steps: constructing a candidate light path through reverse ray tracing, generating a credibility weight in combination with a polarization degree, a reflection frequency, uncertainty and an incident angle index, and normalizing the credibility weight in a local neighborhood; fusing multi-source point clouds under the constraints of weights and anchor points, establishing a parameterized curved surface model, and constructing a joint optimization objective function containing geometric errors and optical consistency errors; and further adaptively adjusting a weight rule based on residual distribution, carrying out closed-loop iterative optimization, and outputting the initial surface shape of the high-reflection curved surface with high precision and good stability.

The invention discloses a pitching side type material taking machine material positioning method based on image recognition. The method comprises the following steps that a structured light sequence is projected to a material stacking area according to a preset coding grating mode; synchronously acquiring multi-view image data of the surface of the stockpile; executing preprocessing; spatial domain processing is carried out, geometric morphology information of the surface of the stockpile is extracted, and height distribution of the surface of the stockpile is analyzed; frequency domain processing is carried out, and light intensity distribution change rule information is extracted; performing fusion processing; time sequence three-dimensional reconstruction operation is executed, and a three-dimensional stockpile surface model is constructed; and determining three-dimensional coordinates of the material target area, and generating material positioning result data. Structured light reconstruction and multi-modal feature fusion are fused, high-precision three-dimensional positioning of the stacked materials is achieved, and the method has the advantages of being precise, robust and high in real-time performance.

A method and a registration device for sampling relevant surface points of a subject for a medical navigation system is presented. From a plurality of surface points of the subject, relevant surface points are accepted and non-relevant surface points are discarded. The method allows the surface geometry acquisition of a subject to be improved. For this purpose, additional information of the surface points is acquired and assigned to the respective surface point for validating each surface point.

The invention discloses a multi-view three-dimensional reconstruction method for a slender object based on curve guidance. According to the method, geometric prior and a deep learning segmentation network are fused, two-stage foreground extraction, a curve-guided motion recovery structure (SfM) and a surface optimization module capable of differential Poisson reconstruction are provided, and a set of complete processing flow is established for the problem that a slender and weak-texture object is difficult to accurately reconstruct. The method comprises the following steps: acquiring a multi-view image through a consumer-level terminal and performing frame extraction to obtain an input data set; generating a high-quality foreground mask by utilizing depth estimation and semantic segmentation; performing curve-guided SfM initialization based on the foreground skeleton curve to realize joint estimation of the camera pose and the sparse three-dimensional curve; carrying out surface geometric optimization through a micro Poisson equation and differentiable grid rendering; and designing a multi-branch loss function fusing luminosity, mask, regularization and Gaussian rendering supervision to train the model. Finally, the method can output a three-dimensional reconstruction result of the slender structure with high geometric accuracy, strong structural integrity and smooth surface, and has good robustness and application value.

In various examples, a deep three-dimensional (3D) conditional generative model is implemented that can synthesize high resolution 3D shapes using simple guides—such as coarse voxels, point clouds, etc.—by marrying implicit and explicit 3D representations into a hybrid 3D representation. The present approach may directly optimize for the reconstructed surface, allowing for the synthesis of finer geometric details with fewer artifacts. The systems and methods described herein may use a deformable tetrahedral grid that encodes a discretized signed distance function (SDF) and a differentiable marching tetrahedral layer that converts the implicit SDF representation to an explicit surface mesh representation. This combination allows joint optimization of the surface geometry and topology as well as generation of the hierarchy of subdivisions using reconstruction and adversarial losses defined explicitly on the surface mesh.

The invention relates to the technical field of three-dimensional modeling, and provides a lightweight hierarchical digital twin modeling and rendering method for a large-scale scene. The method comprises the following steps: analyzing the surface geometry of an original high-precision three-dimensional model, calculating and extracting curvature characteristics of the high-precision three-dimensional model corresponding to each local area, and constructing curvature response distribution through the curvature characteristics; screening and judging each model region corresponding to the original high-precision three-dimensional model based on the obtained curvature response distribution and the dynamic curvature threshold, and obtaining a second key detail region; generating an independent multi-resolution detail enhancement data block for the second key detail area, generating a plurality of data versions with gradually decreased detail degrees for the same second key detail area, and forming a multi-resolution detail library capable of being directly indexed; in the real-time rendering stage, seamless fusion rendering is completed according to the detail data blocks and the basic main body model, and unification of overall scene light weight and local detail high fidelity is achieved.

The invention discloses a method and a system for rapidly evaluating aerodynamic performance of an automobile body appearance. The method specifically comprises the following steps: generating a geometric curved surface sample of an automobile body surface by adopting a parameterized design method; carrying out fluid numerical analysis on a vehicle body external flow field; an encoder for vehicle body appearance surface geometry is established, and the geometric structure of the vehicle body surface is recognized; establishing a prediction model based on a multi-scale map neural network, and performing geometric data memory, transmission and updating on the flow field of the vehicle body appearance surface; and outputting vehicle body aerodynamic characteristic parameters (a resistance coefficient, a lift coefficient and a lateral force coefficient) and an error level. Compared with the prior art, the neural network algorithm is ingeniously designed, rapid prediction of the aerodynamic characteristic parameters of the automobile body is achieved, the method can be used for rapid performance evaluation in the automobile body development process, the design efficiency is improved, and the development cost is reduced.

A surgical system and method for markerless intraoperative navigation is provided. The system can includes a structured light system that can be utilized to intraoperatively sense three-dimensional surface geometry. The computer system is configured to segment a depth image to obtain surface geometry data of an anatomical structure embodied within the depth image, register the surface geometry data of the anatomical structure to a model, and update the surgical plan according to the registered surface geometry data. The surgical system is an endoscopic surgical system.

The application provides an intelligent visual detection method and device for component defects, which comprises the following steps: loading a light standard three-dimensional model corresponding to a target component from a backend server; synchronously collecting real-time image data and depth data of the component on a conveying mechanism or a clamping mechanism; generating a real-time point cloud model for describing surface geometric features of the component based on the real-time image data and the depth data; matching the real-time point cloud model with the light standard three-dimensional model to locate a point area of interest in a high-defect area of the component; spatially mapping and fusing texture features in the real-time image data and the depth data in the point area of interest; and identifying defects in the point area of interest of the component through a joint probability model. The application can realize high-speed and accurate defect detection of components without complex feature engineering, and greatly reduces the maintenance difficulty and implementation cost of the algorithm.

The invention provides a sealing assembly completion state solving method and device, and relates to the technical field of sealing element design. Comprising the following steps: acquiring geometric models of an elastic component and a matched component thereof, constructing an assembly initial configuration, performing finite element discretization on a sealing assembly system, and setting an integral point in the elastic component; in the assembly iteration process, corresponding projection points are determined based on the integration points and the component combination where the integration points are located, and the contact force at the integration points is calculated in combination with material parameters, the positions of the projection points and the surface geometric information of the matched components; and uniformly assembling the contact force and the internal elastic force generated by the elastic component under the current configuration into an overall force balance relationship for nonlinear iterative solution until a preset convergence condition is met, thereby obtaining a sealing assembly completion state. According to the method, the sealing assembly completion state can be stably solved under the condition that initial geometric interference exists, and the convergence and calculation stability of assembly simulation are improved by uniformly processing the contact force and the internal elastic force.

The invention discloses an end-to-end field optimization surface reconstruction method based on Gaussian sputtering guidance, and belongs to the field of computer vision and computer graphics. Comprising the following steps: 1) Gaussian guidance field optimization: defining an implicit symbol distance field by using a parameterized MLP network, and initializing and optimizing a surface through a microcosmic contour surface extraction method; gaussian units are bound with the extracted surface triangular patches, the positions of the patches are parameterized through the vertexes of the patches, rotation and scaling are carried out, regularization constraints based on the minimum wrapped ellipse and the distribution entropy are introduced, it is ensured that the Gaussian units are accurately attached to the surface and evenly distributed on the surface, and a result obtained through rendering guides field optimization. And 2) local multi-resolution: identifying surface voxels based on the optimized symbol distance field to carry out adaptive octree subdivision so as to realize multi-resolution improvement. And 3) adopting an inner and outer double-layer cyclic training architecture, optimizing the symbolic distance field and the grid geometric parameters on the outer layer, and optimizing a bound Gaussian unit on the inner layer to realize end-to-end joint optimization of surface geometry and Gaussian rendering.

This invention relates to the field of image processing technology and discloses a method for target 3D reconstruction and automated model generation. The method includes: acquiring multi-view original point cloud data of the target object, and obtaining a simplified point cloud dataset through voxel filtering and downsampling; dividing the data into smooth regions and sharp edge feature sets based on curvature characteristics, performing smooth fitting and edge-preserving reconstruction respectively, and then generating a composite mesh model through vertex merging and mesh stitching; performing topological structure analysis on the model, detecting and optimizing non-manifold elements and topological holes to obtain an initial mesh model; parametrically mapping and automatically unfolding the texture of the model surface geometric manifold characteristics, and fusing and binding the unfolded texture map with the initial mesh model to finally complete the 3D reconstruction of the target object; this invention can improve the efficiency of target 3D reconstruction and automated model generation.

This invention relates to the field of defect detection technology, specifically a method and system for detecting defects in valve sealing surfaces. The method extracts the position coordinates, geometric contours, and depth parameters of each abnormal morphology region from the three-dimensional surface morphology data of the sealing surface of the valve under inspection; determines the theoretical sealing contact area of ​​the sealing surface in the valve-closed state and spatially maps it with the position coordinates of each abnormal morphology region to filter out the abnormal regions to be judged; calculates the effective contact surface geometry after elastic deformation of the sealing surface under working conditions; based on the depth parameters and effective contact surface geometry of each abnormal region to be judged, if it is determined that the abnormal region to be judged is not completely covered by the effective contact surface after elastic deformation compensation, then the abnormal region to be judged is marked as a functional defect region and the detection result is output; accurately determining whether the abnormal morphology regions on the sealing surface constitute functional defects affecting the sealing function improves the accuracy of valve sealing surface defect detection.

The invention discloses a direct buried pipe foam glass curved surface non-contact type dielectric coating thickness measuring method, and relates to the technical field of non-destructive detection.The direct buried pipe foam glass curved surface non-contact type dielectric coating thickness measuring method comprises the steps that an in-situ calibration reference area is introduced, and a real-time calibration mechanism is established, so that a measuring system can dynamically adjust an internal calculation model according to the actual working condition on site; the absolute precision of a measurement result and the adaptive capacity in a variable field environment are improved, the reliability of each measurement is quantitatively evaluated by generating a credibility index, and a closed-loop remeasurement and data fusion process is established based on the reliability, so that the influence on the judgment of the overall coating quality due to local measurement abnormality is avoided, and the measurement accuracy is improved. According to the invention, the analysis of the geometric morphology of the curved surface is combined with the planning of the measurement path, the intelligent distribution of measurement resources is realized, and the system has the capability of actively coping with the interference of the local surface through the multi-modal dynamic sensing technology.

The application relates to the technical field of image recognition, and discloses a packaging box damage degree online detection method and system based on image recognition, which comprises the following steps: carrying out spatial phase unfolding on a package phase diagram to obtain an absolute phase distribution diagram; generating a surface geometric deformation phase diagram; carrying out spatial alignment on the surface geometric deformation phase diagram to obtain local three-dimensional point cloud data; taking points with a normal vector mutation exceeding a preset threshold value in the local three-dimensional point cloud data as potential damage edge points, and marking points with a maximum principal curvature and minimum principal curvature difference exceeding a preset threshold value as potential deformation points; taking the potential damage edge points as seed points and taking the potential deformation points as growth constraint conditions to iteratively grow in the local three-dimensional point cloud data, and clustering points with continuity and similar geometric mutation characteristics into a damage area; and outputting a damage judgment result according to geometric parameters of the damage area; and the application can improve the efficiency of the packaging box damage degree online detection based on image recognition.

The application discloses a kind of based on image recognition's pitch side formula material positioning method of material handling machine, comprising the following steps: according to preset encoding grating mode, structure light sequence is projected to stacking area;Synchronous acquisition multi-view image data of stacking surface;Execute pre-processing;Carry out spatial domain processing, extract stacking surface geometry information, and analyze stacking surface height distribution;Carry out frequency domain processing, extract light intensity distribution variation law information;Carry out fusion processing;Execute timing three-dimensional reconstruction operation, construct three-dimensional stacking surface model;Determine the three-dimensional coordinates of material target area, generate material positioning result data.The application fuses structure light reconstruction and multimodal feature fusion, realizes stacking material high-precision three-dimensional positioning, with the advantages of accurate, robust and strong real-time.

A method of determining the orientation of a 3D Model to be generated by an additive manufacturing apparatus having a vat for holding photocurable material; and a platform for holding the 3D object corresponding to the 3D model. The platform is relatively movable with respect to the vat and the method includes a step of defining the surface geometry of the 3D model and the surface geometry includes surface segments si.

Techniques for generating a stylized drawing of three-dimensional (3D) shapes using neural networks are disclosed. A processing device generates a set of vector curve paths from a viewpoint of a 3D shape; extracts, using a first neural network of a plurality of neural networks of a machine learning model, surface geometry features of the 3D shape based on geometric properties of surface points of the 3D shape; determines, using a second neural network of the plurality of neural networks of the machine learning model, a set of at least one predicted stroke attribute based on the surface geometry features and a predetermined drawing style; generates, based on the at least one predicted stroke attribute, a set of vector stroke paths corresponding to the set of vector curve paths; and outputs a two-dimensional (2D) stylized stroke drawing of the 3D shape based at least on the set of vector stroke paths.

Computer-implemented methods, computing devices, and articles of manufacture for enhancing video streaming are provided. An example method includes receiving, by a first computing device, a plurality of input images forming the video stream; for each input image of the plurality of input images: applying a geometric model to the input image to determine a surface orientation map indicative of a distribution of illumination on the object in the input image based on a surface geometry of the object; applying an ambient light estimation model to the input image to determine a direction of synthetic illumination to be applied to the input image to enhance at least a portion of the input image; applying a light energy model to determine a quotient image indicative of an amount of light energy to be applied to each pixel of the input image based on the surface orientation map and the direction of the synthetic illumination; and enhancing a portion of the input image based on the quotient image to generate a reilluminated image; and displaying the plurality of reilluminated images as an enhanced video stream.

The invention provides a positioning and mapping method based on a 2D Gaussian surface patch and a related device, and relates to the field of synchronous positioning and mapping. The method comprises the steps of obtaining a to-be-processed image shot by a mobile platform at a current position; wherein the to-be-processed image comprises depth information; updating a currently constructed historical 2D Gaussian map according to the to-be-processed image to obtain a new 2D Gaussian map; and converting the new 2D Gaussian map into a three-dimensional scene map of a scene where the mobile platform is located. Therefore, the 2D Gaussian surface patch is used as scene representation, and the thin structure of the 2D Gaussian surface patch can be used for better aligning the geometric shape of the surface, so that scene surface information is accurately captured, high-precision surface reconstruction can be realized in a low-texture area, and the geometric precision of reconstruction is improved.

The invention relates to an object-level three-dimensional reconstruction method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a sparse point cloud of a target object based on an image set of the target object; dividing a three-dimensional space of the target object through an octree data structure according to the sparse point cloud of the target object, and obtaining signed distance field SDF geometric information in each voxel of an octree according to the sparse point cloud; a plurality of three-dimensional Gaussian distributions are constructed according to the SDF geometric information, iterative optimization is carried out on the plurality of three-dimensional Gaussian distributions, an optimized Gaussian point cloud is obtained, optimization of a loss function is carried out based on characteristic values of a curvature matrix in the optimization process, and the curvature matrix is constructed based on the Gaussian point cloud; and according to the optimized Gaussian point cloud, through the rendered depth map, performing reconstruction to obtain a surface geometric model of the target object. According to the method and the device, high-quality object-level three-dimensional reconstruction can be realized through a reconstruction mode combining an octree and three-dimensional Gaussian splashing.

A method for context-based insertion of a three-dimensional (3D) object into a photograph includes analyzing, by a machine-learning model, a scene of the photograph to determine contextual attributes of the scene. The method also includes selecting, based on the contextual attributes, a 3D asset from a set of 3D assets. The method further includes posing the 3D asset using geometric information determined from the scene such that the posed 3D asset conforms to one or more of perspective, orientation, or surface geometry of the photograph. The method also includes adding a 3D element contextually associated with the scene, the 3D element being selected and positioned in accordance with the contextual attributes. The method further includes generating an output photograph that comprises the posed 3D asset and the 3D elements integrated into the photograph.

A method for detecting deformations of high voltage and / or medium voltage cables and / or cable components, include at a first point of time, capturing and storing (302) a first set of 3-dimensional, 3D, surface geometry measurement data of an area of interest of a surface of the cable or cable component. The method also incudes, at a second point of time, capturing and storing (304) a second set of 3-dimensional, 3D, surface geometry measurement data of an area of interest of a surface of the cable or cable component by moving a 3D surface scanner about the cable over the area of interest. The first and second sets of captured 3D surface geometry measurement data is compared to determine changes that have occurred in the cables or cable components between the first and second points of time, where changes indicate a deformation of the cable or cable component.

The application provides a double-curvature fusion curved surface SLAM navigation method, a longitudinal and transverse double-direction curvature parameter is estimated in real time through double-IMU fusion, an extended information filtering framework containing a double-curvature kinematic model, a laser data curved surface projection compensation algorithm and a path double-curvature compensation algorithm are constructed, and the scheme can accurately depict the curved surface geometric constraint where the robot actually locates; the accumulation of the transverse positioning error is inhibited through online estimation and fusion of the transverse curvature parameter; the accurate description and compensation of the robot side slip effect are realized through the establishment of a kinematic model containing a double-curvature coupling term; the data distortion is eliminated and the environment perception accuracy is improved through the projection of the laser scanning data from a plane to the estimated curved surface; finally, the consistency of the control instruction and the curved surface geometry is ensured through the real-time geometric compensation of the planning path based on the double-curvature parameter, and the stability and trajectory tracking accuracy of the control system under the longitudinal and transverse curvature coupling effect are enhanced.

The application provides a jujube appearance detection and grading method and system based on machine vision, and belongs to the technical field of machine vision and agricultural product quality evaluation. The method comprises the following steps: collecting multi-angle spectral data of jujube samples, combining a three-dimensional reconstruction algorithm to generate a jujube shape field model; extracting surface geometric features of the shape field model to obtain a three-dimensional shape feature vector set; synchronously collecting microtexture spectral data and chrominance information of the jujube samples, performing nonlinear chrominance transformation processing to obtain a microtexture spectral feature vector set and a nonlinear chrominance feature vector set; and through a multi-view hyperspectral imaging system and a three-dimensional reconstruction algorithm, spectral and shape data of the jujube samples can be efficiently and accurately collected and processed, so that the efficiency and accuracy of grading are greatly improved.

This invention relates to the field of precision inspection and discloses a defect detection method based on machine vision. The method includes: an illumination device projecting a modulated light field containing multiple modulation periods, wherein the spatial light intensity distribution is shifted at a preset speed between adjacent periods; a photoelectric detection array acquiring a time-varying light intensity sequence reflected from the material surface, the sequence containing the reflected light intensity values ​​of multiple sampling points within the corresponding period; extracting the transient change gradient of the reflected light intensity within adjacent periods for sampling points at the same physical coordinate position on the corresponding material surface; determining an intrinsic light intensity change threshold based on the shift speed and sampling frequency; and identifying sampling points whose transient change gradient exceeds the threshold as physical breakage defects. This invention utilizes the constraint law of the geometric continuity of the material surface on the optical response to achieve the physical separation of texture interference signals and defect abrupt change signals, thereby solving the problem of misjudgment in defect measurement under complex topological surfaces.

The invention relates to an ancient ceramic three-dimensional reconstruction and incomplete part virtual repair method and device, equipment and a medium. According to the method, surface grids and internal structure feature points of CT data are extracted, registration with optical surface grids is carried out, spectrogram matching is carried out by using the internal feature points to obtain initial alignment, and a CT descriptor and an optical descriptor fusing internal structure information are constructed; through joint optimization of surface geometry, descriptor similarity and internal topology consistency, high-precision registration is obtained, and a complete three-dimensional model fused with an internal structure is generated; and finally, segmenting the incomplete region in combination with multi-modal verification, and guiding the generation of a repair curved surface by using an associated internal structure feature field. According to the method, an internal structure is introduced as a core matching constraint, so that the problem of registration failure caused by surface information deviation and feature sparseness in a traditional method is solved, a subsequent virtual repair result simultaneously meets the geometric smoothness of the appearance and the distribution consistency of an internal material, and the integrity of a cultural relic digital model and the scientificity of repair are improved.